CN1306395C - Processor extended instruction of MIPS instruction set, encoding method and component thereof - Google Patents
Processor extended instruction of MIPS instruction set, encoding method and component thereof Download PDFInfo
- Publication number
- CN1306395C CN1306395C CNB2004100394601A CN200410039460A CN1306395C CN 1306395 C CN1306395 C CN 1306395C CN B2004100394601 A CNB2004100394601 A CN B2004100394601A CN 200410039460 A CN200410039460 A CN 200410039460A CN 1306395 C CN1306395 C CN 1306395C
- Authority
- CN
- China
- Prior art keywords
- instruction
- extended
- point
- floating point
- floating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Lifetime
Links
- 238000000034 method Methods 0.000 title claims abstract description 14
- 230000014759 maintenance of location Effects 0.000 claims description 13
- 230000014509 gene expression Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
Images
Landscapes
- Advance Control (AREA)
Abstract
本发明公开了一种MIPS指令集的处理器扩展指令及其编码方法和部件。该扩展指令编码成MIPS指令集中浮点指令的格式，该格式中包括一格式域；所述扩展指令的格式域取值为现有的MIPS指令集中的浮点指令的格式域的保留值。该扩展指令的执行部件为一种功能增强的浮点部件，包括数据通路、输入寄存器、输出寄存器和执行浮点操作的浮点指令执行模块，还包括一个与所述浮点指令执行模块并联的扩展指令执行模块，一个判断指令是浮点指令还是扩展指令的选择模块。本发明在指令扩展时避免了引入新的指令格式而增加指令译码的复杂度，扩展指令复用原有的浮点指令的数据通路以及寄存器资源，避免了芯片复杂度的扩大，大大节省了芯片的面积。
The invention discloses a processor extension instruction of a MIPS instruction set, a coding method and components thereof. The extended instruction is encoded into the format of the floating-point instruction in the MIPS instruction set, and the format includes a format field; the value of the format field of the extended instruction is a reserved value of the format field of the floating-point instruction in the existing MIPS instruction set. The execution unit of the extended instruction is a function-enhanced floating-point unit, which includes a data path, an input register, an output register, and a floating-point instruction execution module for performing floating-point operations, and also includes a parallel connection with the floating-point instruction execution module. The extended instruction execution module is a selection module for judging whether the instruction is a floating-point instruction or an extended instruction. The present invention avoids the introduction of a new instruction format to increase the complexity of instruction decoding when the instruction is extended, and the extended instruction reuses the data path and register resources of the original floating-point instruction, avoiding the expansion of the chip complexity and greatly saving area of the chip.
Description
Technical field
The present invention relates to the MIPS instruction set, in particular, the present invention relates to processor extended instruction and the coding method and the execution unit of MIPS instruction set.
Background technology
The MIPS instruction set is a kind of reduced instruction set computer, has the characteristic of 32 of every instruction fixed length.The MIPS instruction set generally will comprise the instruction of types such as floating point instruction, fixed point instruction and multimedia instruction, to realize dissimilar operations.
In traditional realization to the MIPS instruction set, different instruction types is carried out by different execution units.For example, in realization, to comprise floating point unit, fixed point parts and multimedia components at least usually to the MIPS instruction set.And different execution units also needs data path and register file for this execution unit design specialized except its instruction execution module, and for example multimedia components and fixed point parts all have own special-purpose data path and register file respectively.
In the MIPS instruction set, the 25th to the 21st (one of the rightmost of instruction is that lowest order is the 0th) in the floating point instruction form (see figure 1) is form (fmt) territory of floating point instruction, decide the object type of floating-point operation type and operation by this territory, for example fmt equal 16 expression operations to as if single precision floating datum, fmt equal 17 expression operations to as if double-precision floating points etc.0 to 31 totally 32 different values can be represented in five fmt territory, have used 11 values such as 0,1,2,4,5,6,8,16,17,20,21 in the MIPS instruction set, and other value is as retention.
Summary of the invention
The objective of the invention is to utilize floating point instruction in the MIPS instruction set the fmt territory retention for the MIPS instruction set provide a kind of extended instruction and coding method, the expansion fixed point instruction that this extended instruction comprises the extended multimedia instruction that realizes multimedia operations and/or realizes fixed-point operation.Another object of the present invention provides a kind of execution unit for this extended instruction, and these parts are the floating point unit of increased functionality.
To achieve these goals, the invention provides a kind of processor extended instruction, the form of described extended instruction is identical with the form of floating point instruction in the MIPS instruction set; Described extended instruction comprises a format field; The format field value of described extended instruction is the retention of the format field of the floating point instruction in the existing MIPS instruction set.
Described extended instruction comprises the extended multimedia instruction that realizes the multimedia computing.The instruction of described extended multimedia realizes the function of existing MIPS instruction concentrated part or whole multimedia instruction.
Described extended instruction comprises the expansion fixed point instruction that realizes fixed-point arithmetic.The instruction of described expansion fixed point realizes the function of the fixed point instruction of existing MIPS instruction concentrated part.
The present invention also provides a kind of coding method of processor extended instruction, is used to existing MIPS instruction set that extended instruction is provided, and described extended instruction is encoded into the form of floating point instruction in the MIPS instruction set, comprises a format field in this form; The format field value of described extended instruction is the retention of the format field of the floating point instruction in the existing MIPS instruction set.
Described extended instruction comprises the extended multimedia instruction that is used to realize the multimedia computing and/or is used to realize the expansion fixed point instruction of fixed-point arithmetic.
In order to carry out above-mentioned extended instruction, the present invention also provides a kind of floating point unit of increased functionality, comprises the floating point instruction execution module of data path, input register, output register and execution floating-point operation, also comprises:
An extended instruction execution module in parallel with described floating point instruction execution module is used to carry out extended instruction;
Select module for one, the value decision instruction that is used for according to the instruction format field is floating point instruction or extended instruction.
When extended instruction is that described extended instruction execution module then is the multimedia instruction execution module when carrying out the instruction of multimedia operations.
When extended instruction was the instruction of execution fixed-point operation, 14 of described extended instruction execution modules were fixed point instruction execution module.
Advantage of the present invention mainly is to avoid introducing new order format and the complexity that increases instruction decode, keeps the original instruction style of MIPS instruction set, makes instruction decode and originally consistent.Have only fmt territory difference simultaneously, make new extended instruction also be taken as a kind of new floating point instruction, the data path of multiplexing original floating point instruction and register resources have been avoided the expansion of chip complexity, have saved area of chip greatly.Further, the multimedia instruction of expansion provided by the invention can replace the multimedia instruction in original MIPS instruction set fully, thereby when realizing, saved by processor multimedia components and data path and vector register file, greatly the complexity of simplified processor chip and chip area.
Description of drawings
Fig. 1 is the floating point instruction form synoptic diagram of MIPS instruction set;
Fig. 2 is that the present invention utilizes MIPS floating point instruction fmt territory extended instruction and multiplexing floating point unit to carry out the process flow diagram of extended instruction;
Fig. 3 is a floating point unit synoptic diagram of the prior art;
Fig. 4 is the synoptic diagram of an embodiment of floating point unit of increased functionality of the present invention;
Fig. 5 is the synoptic diagram of another embodiment of floating point unit of increased functionality of the present invention;
Fig. 6 is the coding according to the multimedia instruction PADDB of the present invention's expansion;
Fig. 7 is the coding according to the fixed point instruction Dsub of the present invention's expansion;
Fig. 8 is the order format of the multimedia instruction in the existing MIPS instruction set.
Embodiment
Below in conjunction with the drawings and specific embodiments the present invention is described in further detail.
The floating point instruction of MIPS instruction set as shown in Figure 1, its length is 32, comprises six territories;
The 31st～26 operation code field (opcode territory), for floating point instruction, the value in opcode territory is fixed as 010001, waits other instruction with the instruction of difference fixed point;
The 25th～21 format field (fmt territory), fmt territory are used for representing that operation result is that single precision floating datum, double-precision floating points, 32 integers, 64 integers or expression are floating-point transfer instructions etc.;
Second source flating point register number represented in the 20th～16 source flating point register 2 territories (ft territory);
First source flating point register number is represented in the 15th～11 source flating point register 1 territory (fs territory);
The 10th～6 the territory of flating point register as a result (fd territory), ecbatic flating point register number;
The 5th～0 domain-functionalities territory (function), the function of expression computing, for example addition subtraction multiplication and division etc.
Wherein, 0 to 31 totally 32 different values can be represented in five fmt territory, have used 11 values such as 0,1,2,4,5,6,8,16,17,20,21 in the MIPS instruction set, and other value is as retention.
The present invention utilizes the retention in the floating point instruction to realize extended instruction, the expansion fixed point instruction that this extended instruction comprises the extended multimedia instruction that realizes multimedia operations and/or realizes fixed-point operation.When needs expand new instruction for the MIPS instruction set, at first to determine the coding of new instruction, the method applied in the present invention is exactly to apply mechanically the form of the floating point instruction of MIPS instruction set, is only made amendment in the fmt territory in the floating point instruction.Below in conjunction with the process flow diagram of Fig. 2 concrete implementation process that makes an explanation.
As shown in Figure 2, in the step 101, at first select a floating point instruction in the existing MIPS instruction set.
In the step 102, change the value in the fmt territory of selected floating point instruction, become the coding of new extended instruction.The fmt territory value of new extended instruction is the retention in the fmt territory of existing floating point instruction, and is different with original instruction with expression by the difference of fmt territory value.In new extended instruction coding, the meaning of removing other overseas domain representation of fmt is identical with original floating point instruction.Like this, for extended instruction provided by the invention, its form is identical with the form of floating point instruction shown in Figure 1, and difference is that the fmt territory of this extended instruction adopted obsolete retention in the existing floating point instruction.Further, though the opcode territory value in this extended instruction is 010001, should instruct not therefore and necessarily as floating point instruction.In fact, the instruction type of this extended instruction (as types such as floating point instruction, fixed point instruction or multimedia instructions) is not to judge by the opcode territory in the instruction, but judges by the value in fmt territory.In this extended instruction, the opcode territory is changed to 010001 of expression floating-point, just for the resource in the multiplexing existing floating point unit when carrying out this extended instruction, this point is with clearer understanding in the following description.
Particularly in the step 103 in Fig. 2, because extended instruction of the present invention is revised original floating point instruction and is formed, therefore make extended instruction have the order format identical with floating point instruction, extended instruction provided by the invention just may be carried out at original floating point unit that is used for carrying out floating point instruction.Like this, just do not need, do not need so just to have avoided the expansion of chip complexity yet, saved area of chip greatly for these parts provide special data path and register file for new extended instruction provides corresponding execution unit specially.
After adopting extended instruction coding of the present invention, in instruction decode, because the value in the opcode territory of new extended instruction is still represented 010001 of floating point instruction, utilizing the original value with the opcode territory is the decoding mechanism that 010001 instruction is translated into floating point instruction, therefore still extended instruction is decoded into a kind of floating point instruction.Like this, in the processor decoding unit just need not revise, identical with decoding unit of the prior art.By the data path of original floating point instruction, floating point unit is delivered in new instruction then, this part processor that therefore relates to data path is realized also need not revising.In the present invention, only need to strengthen the processing power of floating point unit, make it can the decision instruction type, so that carry out different operations.
Fig. 3 shows the composition synoptic diagram of existing floating point unit, comprises the data path of representing with arrow 10, the input register 11 that comprises floating point unit and output register 12, and a floating point instruction execution module 13 that is used to carry out floating-point operation.Floating point instruction is handled in floating point instruction execution module 13 after sending into floating point unit by input register 11, at last the result is sent floating point unit by output register 12 and writes back flating point register heap (not shown among Fig. 3).Such floating point unit only can be carried out existing floating point instruction.
In order to strengthen the processing capacity of floating point unit, the invention provides the floating point unit of increased functionality as shown in Figure 4.Data path 10 in the still multiplexing existing floating point unit of the floating point unit of this increased functionality comprises the input register 11 and the output register 12 of floating point unit, and the floating point instruction execution module 13 of carrying out floating-point operation.Simultaneously, this floating point unit provides an extended instruction execution module 14 that is in parallel with floating point instruction execution module 13 on its data path 10, so that carry out corresponding extended instruction.This floating point unit has comprised that is also selected a module 15, is used for the type according to the value decision instruction in the fmt territory of instructing, so that floating point instruction execution module 13 or 14 processing of extended instruction execution module are sent in instruction.Article one, after instruction enters input register 11, by selecting module 15 to judge that according to the value in its fmt territory this instruction is floating point instruction or has the extended instruction of floating point instruction form.When this instruction is floating point instruction, it is sent into processing in the floating point instruction execution module 13, send the floating point unit of this increased functionality then by output register 12, write back flating point register heap (not shown among Fig. 4); When this instruction is when having the extended instruction of floating point instruction form, it is sent in the extended instruction execution module 14 handle, send the floating point unit of this increased functionality then by output register 12, write back flating point register heap (not shown among Fig. 4).When extended instruction is that 14 of the extended instruction execution modules among Fig. 4 are the multimedia instruction execution module when carrying out the instruction of multimedia operations; When extended instruction was the instruction of execution fixed-point operation, 14 of the extended instruction execution modules among Fig. 4 were fixed point instruction execution module, by that analogy.
Another embodiment of the floating point unit of increased functionality as shown in Figure 5, its basic composition is identical with Fig. 4, difference is, after floating point instruction is sent into floating point unit, in floating point instruction execution module 13 and extended instruction execution module 14, handle simultaneously, select module 15 to select correct result to send floating point unit then according to the value in fmt territory.
Next will enumerate two specific embodiments, the fixed point of the multimedia instruction of an expansion and an expansion is instructed and is specified the method for how mentioning by the present invention and come extended instruction coding and multiplexing floating point unit are carried out.
Example 1: increase a multimedia vector and add instruction PADDB, this instruction is with eight the 8 figure places addition respectively in two source operands.At first select an existing floating add instruction, the value of inciting somebody to action fmt territory wherein then is changed to a retention, and the fmt territory is set to 11110 in this example, it is formed new coding, as shown in Figure 6.That is to say that in instruction shown in Figure 6, the multimedia vector of value 11110 expressions in the fmt territory of this instruction adds instruction PADDB.The meaning in other territory is identical with the instruction of original floating add, and the value in opcode territory is that 010001 expression is a floating point instruction; First source flating point register of fs domain representation number; Second source flating point register of ft domain representation number; The fd domain representation is flating point register number as a result; The value in function territory is 000000 expression additive operation.When processor is realized, extended instruction execution module 14 in the floating point unit of Fig. 4 or increased functionality shown in Figure 5 is the multimedia instruction execution module, selection module 15 is wherein judged its multimedia instruction for expansion according to the value 11110 in fmt territory, and that floating point unit is exported is the result that multimedia execution module is handled this instruction.
Example 2: increase a fixed point subtraction instruction Dsub, this instruction is subtracted each other two 64 integers exactly.At first select an existing floating-point subtraction instruction, the value in fmt territory wherein is changed to a retention then, forms new coding, and the fmt territory is set to 01111 in this example, it is formed new coding, as shown in Figure 7.That is to say, in instruction shown in Figure 7, the value 01111 expression fixed point subtraction instruction Dsub in the fmt territory of this instruction.The meaning in other territory is identical with original floating-point subtraction instruction, and the value in opcode territory is that 010001 expression is a floating point instruction; First source flating point register of fs domain representation number; Second source flating point register of ft domain representation number; The fd domain representation is flating point register number as a result; The value in function territory is 000001 expression subtraction.When processor is realized, extended instruction execution module 14 in the floating point unit of Fig. 4 or increased functionality shown in Figure 5 is fixed point instruction execution module, selection module 15 is wherein judged its fixed point instruction for expansion according to the value 01111 in fmt territory, and that floating point unit is exported is the result that fixed point instruction execution module is handled this instruction.
In existing MIPS instruction set, the multimedia instruction of MIPS uses and the diverse coding of floating point instruction, as shown in Figure 8.Traditional realization is to carry out multimedia instruction with a new multimedia components, the weak point of this realization is data path and the vector register file that needs to give the multimedia components design specialized, and this can increase considerably the complexity and the chip area of chip design.By analyzing existing multimedia instruction as shown in Figure 8, as seen its form is more regular, and closely similar with existing floating point instruction shown in Figure 1, difference is: the opcode territory is that 010010 expression is a multimedia instruction; Sel domain representation operand is 8,16,32 or 64 s' a vector; First source vector register number of vs domain representation; Second source vector register number of vt domain representation; Vd domain representation result vector register number; The vector operation that the function domain representation is different.According to the format character of existing multimedia instruction shown in Figure 8, can find that adopting after the following measure is that the multimedia instruction of available expansion replaces original multimedia instruction in the MIPS instruction set:
(1) with the value change in the opcode territory of multimedia instruction, is changed to 010001 of expression floating point instruction;
(2) with the fmt territory in the floating point instruction of corresponding in the sel territory in the multimedia instruction, and with its value is the retention in fmt territory.
To add instruction PADDB (Fig. 6) promptly be that many matchmakers with expansion instruct an example of the multimedia instruction of replacement in the existing MIPS instruction set to the multimedia vector of example 1 described expansion in the preamble.
Like this, when processor is realized, just original multimedia instruction can be decoded as floating point instruction, and in floating point unit, carry out, and judge its instruction type according to the value in fmt territory (the sel territory in the former in other words multimedia instruction) and meet at multimedia execution module by the selection module 15 in the floating point unit of increased functionality and carry out as Fig. 4 or increased functionality shown in Figure 5.Multimedia instruction has just saved original multimedia components and data path and register resources in the processor when processor is realized like this, and realize multimedia operations, thereby the complexity and the chip area of chip design have been reduced with the multimedia instruction of expansion and the floating point unit of increased functionality.
Claims (10)
1, a kind of processor extended instruction is characterized in that, the form of described extended instruction is identical with the form of floating point instruction in the MIPS instruction set; Described extended instruction comprises a format field; The format field value of described extended instruction is the retention of the format field of the floating point instruction in the existing MIPS instruction set.
2, processor extended instruction according to claim 1 is characterized in that, described extended instruction comprises the extended multimedia instruction that realizes the multimedia computing.
3, processor extended instruction according to claim 2 is characterized in that, the instruction of described extended multimedia realizes the function of existing MIPS instruction concentrated part or whole multimedia instruction.
4, processor extended instruction according to claim 1 is characterized in that, described extended instruction comprises the expansion fixed point instruction that realizes fixed-point arithmetic.
5, processor extended instruction according to claim 4 is characterized in that, the instruction of described expansion fixed point realizes the function of the fixed point instruction of existing MIPS instruction concentrated part.
6, a kind of coding method of processor extended instruction is used to existing MIPS instruction set that extended instruction is provided, and it is characterized in that described extended instruction is encoded into the form of floating point instruction in the MIPS instruction set, comprises a format field in this form; The format field value of described extended instruction is the retention of the format field of the floating point instruction in the existing MIPS instruction set.
7, the coding method of processor extended instruction according to claim 6 is characterized in that, described extended instruction comprises the extended multimedia instruction that is used to realize the multimedia computing and/or is used to realize the expansion fixed point instruction of fixed-point arithmetic.
8, a kind of floating point unit of increased functionality comprises it is characterized in that the floating point instruction execution module of data path, input register, output register and execution floating-point operation, also comprises:
An extended instruction execution module in parallel with described floating point instruction execution module is used to carry out extended instruction;
Select module for one, the value decision instruction that is used for according to the instruction format field is floating point instruction or extended instruction.
9, the floating point unit of increased functionality according to claim 8 is characterized in that, when extended instruction is that described extended instruction execution module then is the multimedia instruction execution module when carrying out the instruction of multimedia operations.
10, the floating point unit of increased functionality according to claim 8 is characterized in that, when extended instruction was the instruction of execution fixed-point operation, described extended instruction execution module (14) then was fixed point instruction execution module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100394601A CN1306395C (en) | 2004-02-13 | 2004-02-13 | Processor extended instruction of MIPS instruction set, encoding method and component thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2004100394601A CN1306395C (en) | 2004-02-13 | 2004-02-13 | Processor extended instruction of MIPS instruction set, encoding method and component thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1655117A CN1655117A (en) | 2005-08-17 |
| CN1306395C true CN1306395C (en) | 2007-03-21 |
Family
ID=34892352
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2004100394601A Expired - Lifetime CN1306395C (en) | 2004-02-13 | 2004-02-13 | Processor extended instruction of MIPS instruction set, encoding method and component thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN1306395C (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101216755B (en) * | 2008-01-02 | 2011-10-12 | 中国科学院计算技术研究所 | RISC method and its floating-point register non-alignment access method |
| TWI601062B (en) * | 2015-10-08 | 2017-10-01 | 上海兆芯集成電路有限公司 | Apparatus employing user-specified binary point fixed point arithmetic |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104615443B (en) * | 2015-02-27 | 2018-04-24 | 珠海市杰理科技股份有限公司 | MCU Instruction extended method and system |
| CN108415728B (en) * | 2018-03-01 | 2020-12-29 | 中国科学院计算技术研究所 | A method and device for executing extended floating-point arithmetic instructions for a processor |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1303501A (en) * | 1998-05-27 | 2001-07-11 | Arm有限公司 | Mixed vector/scalar register file |
| WO2003088071A2 (en) * | 2002-04-10 | 2003-10-23 | Tensilica, Inc. | High-performance hybrid processor with configurable execution units |
| JP3493369B2 (en) * | 1994-12-13 | 2004-02-03 | 株式会社ルネサステクノロジ | Computer |
-
2004
- 2004-02-13 CN CNB2004100394601A patent/CN1306395C/en not_active Expired - Lifetime
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3493369B2 (en) * | 1994-12-13 | 2004-02-03 | 株式会社ルネサステクノロジ | Computer |
| CN1303501A (en) * | 1998-05-27 | 2001-07-11 | Arm有限公司 | Mixed vector/scalar register file |
| WO2003088071A2 (en) * | 2002-04-10 | 2003-10-23 | Tensilica, Inc. | High-performance hybrid processor with configurable execution units |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101216755B (en) * | 2008-01-02 | 2011-10-12 | 中国科学院计算技术研究所 | RISC method and its floating-point register non-alignment access method |
| TWI601062B (en) * | 2015-10-08 | 2017-10-01 | 上海兆芯集成電路有限公司 | Apparatus employing user-specified binary point fixed point arithmetic |
Also Published As
| Publication number | Publication date |
|---|---|
| CN1655117A (en) | 2005-08-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9164763B2 (en) | Single instruction group information processing apparatus for dynamically performing transient processing associated with a repeat instruction | |
| RU2273044C2 (en) | Method and device for parallel conjunction of data with shift to the right | |
| CN1126034C (en) | Emulation of instruction set on instruction set architecture transition | |
| CN100570552C (en) | A parallel floating-point multiply-accumulate unit | |
| CN1174313C (en) | Multiple instruction set data process | |
| CN1250906A (en) | Use composite data processor systemand instruction system | |
| CN1402843A (en) | Processing multiply-accumulate operations in single cycle | |
| CN1735880A (en) | Method and system for performing calculation operations and a device | |
| CN1421001A (en) | Optimization of N-base typed arithmetic expressions | |
| CN1215862A (en) | Computing method and computing device | |
| CN1577257A (en) | SIMD integer multiply high with round and shift | |
| CN1613054A (en) | Controlling compatibility levels of binary translations between instruction set architectures | |
| CN1193771A (en) | Processor suitable for trimming processing consisting of positive value processing and saturation operation processing | |
| CN1503936A (en) | Partial bitwise permutation | |
| CN1672132A (en) | Source-to-source partitioning compilation | |
| CN101042640A (en) | Digital signal processor with bit expansion and bit compressing compressing cell | |
| CN1165838C (en) | Method and processor for storing packed operands in a register file | |
| CN1216327C (en) | 32-bit Embedded Microprocessor Using Dual Instruction Sets | |
| CN1185606A (en) | Single Instruction Multiple Data Correction Circuit for Arithmetic/Shift Operations | |
| CN1306395C (en) | Processor extended instruction of MIPS instruction set, encoding method and component thereof | |
| KR20210122828A (en) | Encoding Special Values in Anchor Data Elements | |
| CN1560731A (en) | 32-bit media digital signal processor | |
| CN1508674A (en) | Method for providing width-variable at least six-path addition instruction and apparatus thereof | |
| CN1512324A (en) | Extended precision integer divide algorithm | |
| CN1127264C (en) | Hierarchy programmable parallel video signal processor structure for motion estimation algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Assignee: Beijing Loongson Technology Service Center Co.,Ltd. Assignor: Institute of Computing Technology, Chinese Academy of Sciences Contract fulfillment period: 2009.12.16 to 2028.12.31 Contract record no.: 2010990000062 Denomination of invention: Processor extended instruction of MIPS instruction set, encoding method and component thereof Granted publication date: 20070321 License type: exclusive license Record date: 20100128 |
|
| LIC | Patent licence contract for exploitation submitted for record |
Free format text: EXCLUSIVE LICENSE; TIME LIMIT OF IMPLEMENTING CONTACT: 2009.12.16 TO 2028.12.31; CHANGE OF CONTRACT Name of requester: BEIJING LOONGSON TECHNOLOGY SERVICE CENTER CO., LT Effective date: 20100128 |
|
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: LOONGSON TECHNOLOGY Corp.,Ltd. Assignor: Institute of Computing Technology, Chinese Academy of Sciences Contract record no.: 2010990000062 Date of cancellation: 20141231 |
|
| EM01 | Change of recordation of patent licensing contract |
Change date: 20141231 Contract record no.: 2010990000062 Assignee after: LOONGSON TECHNOLOGY Corp.,Ltd. Assignee before: Beijing Loongson Technology Service Center Co.,Ltd. |
|
| LICC | Enforcement, change and cancellation of record of contracts on the licence for exploitation of a patent or utility model | ||
| EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20050817 Assignee: LOONGSON TECHNOLOGY Corp.,Ltd. Assignor: Institute of Computing Technology, Chinese Academy of Sciences Contract record no.: 2015990000066 Denomination of invention: Processor extended instruction of MIPS instruction set, encoding method and component thereof Granted publication date: 20070321 License type: Common License Record date: 20150211 |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200824 Address after: 100095, Beijing, Zhongguancun Haidian District environmental science and technology demonstration park, Liuzhou Industrial Park, No. 2 building Patentee after: LOONGSON TECHNOLOGY Corp.,Ltd. Address before: 100080 Haidian District, Zhongguancun Academy of Sciences, South Road, No. 6, No. Patentee before: Institute of Computing Technology, Chinese Academy of Sciences |
|
| EC01 | Cancellation of recordation of patent licensing contract | ||
| EC01 | Cancellation of recordation of patent licensing contract |
Assignee: LOONGSON TECHNOLOGY Corp.,Ltd. Assignor: Institute of Computing Technology, Chinese Academy of Sciences Contract record no.: 2015990000066 Date of cancellation: 20200928 |
|
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: 100095 Building 2, Longxin Industrial Park, Zhongguancun environmental protection technology demonstration park, Haidian District, Beijing Patentee after: Loongson Zhongke Technology Co.,Ltd. Address before: 100095 Building 2, Longxin Industrial Park, Zhongguancun environmental protection technology demonstration park, Haidian District, Beijing Patentee before: LOONGSON TECHNOLOGY Corp.,Ltd. |
|
| CX01 | Expiry of patent term |
Granted publication date: 20070321 |
|
| CX01 | Expiry of patent term |